Especially towards the end of the experiment, biomass in the W treatment showed a trend of becoming higher than in the other two treatments, but a significant difference was only found between W and H treatments (KW test: chi 2 = 7.65, df = 2, p = .02, W‐H: p < .01) at the end of the second heatwave (day 21, Table S2 ).
We found an increasing trend in the ZP:PP biomass ratio with significant differences between W and H at the end of the second heatwave (day 21; KW test: chi 2 = 4.77, df = 2, p = .09, H‐W: p = .03, Figure 1e , Table S2 ).
As Chl a still showed a decreasing trend during the first 4 days (days 0–3) in all treatments, likely linked to filling up the mesocosms (e.g., ongoing sedimentation), we excluded this period from the analysis.